Outcome of patients with advanced gastro-intestinal stromal tumors (GIST) crossing over to a daily imatinib dose of 800mg (HD) after progression on 400 mg (LD)- an international, intergroup study of the EORTC, ISG and AGITG

Reviewer: Neha Vapiwala, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 6, 2004

This presentation discusses the use of imatinib at a dose that has not been approved by the FDA.

Presenter: J.R. Zalcberg
Presenter's Affiliation: EORTC/ ISG/AGITG
Type of Session: Scientific


This international intergroup has previously demonstrated an improvement in progression-free survival (PFS) with the use of high-dose (HD) imatinib in patients with gastrointestinal stromal tumors (GIST), as compared to those treated with low-dose (LD) imatinib. In that trial, patients on the LD arm were permitted to crossover to the HD arm in the event of disease progression during LD treatment. This study was designed to specifically evaluate the safety and efficacy of allowing this "dose escalation" crossover in patients with progressive disease, rather than declaring treatment failure and abandoning imatinib in favor of another therapy.

Materials and Methods

  • 946 patients with advanced GIST were enrolled between 2/13/01 and 2/8/02, from 13 countries and 56 centers
  • Randomization to LD imatinib at 400 mg daily (n= 473) vs. HD imatinib at 800 mg daily (n= 473) 
  • Primary areas of interest in this study were as follows:
    1. proportion of patients on LD arm that crossed over to HD arm
    2. the cumulative incidence of dose reductions needed after crossover to the HD arm
    3. the influence, if any, of dose reductions needed prior to the crossover
    4. the toxicity profiles of crossover patients, comparing adverse effects during the LD treatment to    those occurring on HD imatinib
  • The Growth Modulation Index (GMI) first proposed by Mick et al. was employed here. This tool is used for evaluating if one agent is more active than another. It is calculated by comparing the time to progression on a study agent to the time to progression on a standard therapy. If the ratio is >1.33, it is believed that the study agent is "more active" than the standard.


  • Of the 473 patients initially randomized to LD imatinib, 241 had progressive disease (PD) on treatment
  • Of the 241 with PD, 133 (55%) crossed over to the HD arm according to protocol
  • There were fewer dose reductions needed in the crossover group of patients, compared to the number of reductions needed in the patients originally randomized to HD therapy.
  • Of the 133 patients who crossed over to the HD arm, 11 (8%) had dose reductions before developing  PD and crossing over, and 31% of the required dose reductions after crossover occurred in this group.
  • Of the 133 patients who crossed over to the HD arm, 122 (92%) had no dose reductions before developing  PD and crossing over, and 69% of the dose reductions after crossover occurred in this group. 
  • Fatigue was more severe in patients after the crossover to HD compared to before the crossover (47% vs. 21%, p= 0.002).
  • Anemia was more severe in patients after the crossover to HD compared to before the crossover (51% vs. 15%, p= 0.00001).
  • Neutropenia was less severe in patients after the crossover to HD compared to before the crossover (13% vs. 30%, p= 0.015).
  • Of 97 evaluable patients among the 133 who crossed over, there were 2.5% with confirmed partial response and 30.3% with stable disease, as measured by the RECIST criteria.
  • The PFS rate at 1 year for the 97 evaluable patients was 18.1%, with a median time to progresison of 81 days.
  • The GMI suggested greater activity of the HD imatinib in at least 24.5% of patients.
  • The median time patients remained on the HD treatment after crossover was 4 months, with 23% of the crossover group still on HD imatinib at 1 year.


Author's Conclusions

  • This large intergroup study evaluated the effects of crossover from LD to HD in 55% of patients with progressive disease on the LD arm.
  • The vast majority of these patients (92%) had no history of dose reductions prior to the crossover.
  • A history of dose reductions prior to the crossover did not appear to predict for dose reductions after the crossover.
  • Patients who crossed over to the HD therapy were less likely to have dose reductions than those originally assigned to the HD therapy. 
  • Among the crossover patients, anemia and fatigue were worse on the HD therapy compared to when they were on the LD treatment, while neutropenia was better on the HD than the LD.
  • The total response rate was 32.8%, comprised almost entirely of stable disease.
  • The majority of patients discontinued the HD therapy by one year after the crossover due to progressive disease, with a PFS at 1 year of 18.1%.


Clinical/Scientific Implications

This important study presents preliminary data suggesting that some advanced GIST patients with progressive disease on LD imatinib therapy may benefit from dose escalation to HD imatinib. Although the majority of patients continued to have disease progression after crossing over to HD therapy, there was a response rate of 32.8% and a one-year PFS rate of 18.1%, showing some benefit to the 800 mg dose even after the 400mg dose did not succeed. Furthermore, the GMI calculation lends support  to this practice, with an index >1.33 in 24.5% indicating greater activity of HD imatinib compared to LD imatinib. The authors admit that the application of the GMI is hypothesis-generating, but does prompt further research in this area. Overall, this trial does support the practice of attempting higher dose imatinib treatment in patients who fail the low-dose therapy, rather than immediately discontinuing imatinib and switching to a different drug.

Two possible mechanisms might explain the benefit of HD therapy shown here:

  1. use of the HD imatinib may overcome tolerance or inherent resistance in certain patients
  2. use of the HD imatinib leads to improved metabolism/pharmocodyanics compared to LD

What might be useful in this study is a measurement of the drug serum levels of imatinib in the crossover patients before and after the crossover from LD to HD. This might help further clarify the mechanism behind the observed effect.

Finally, it is not surprising that certain adverse effects, such as anemia and fatigue, would be worse on the HD therapy compared to LD, but it is not clear why the neutropenia should improve. This finding raises the possibility that some imatinib-related toxicities may in fact decrease with increasing duration of treament.

Oncolink's ASCO Coverage made possible by an unrestricted Educational Grant from Bristol-Myers Squibb Oncology.